Ouroboros DeFi "O-De"

Play the audio

@mikeysmith @emma_j_saunders @grok was does trumps post mean in the screenshot ?

God, I hate being right all the time

Delete every prompt template you ever saved. You only need these 3 files: 1. Folder ☑ A folder on your computer called "Cowork": This is where your 3 files live, forever. ↳ Claude reads them before every single task. 2. about-me .md ☑ Who you are. How you think. How you write: Open Claude. Run a 100-question interview. Then compress the answers into one .md file. ↳ Now Claude sounds exactly like you. 3. anti-ai-writing-style .md ☑ Every word, pattern, and tone you HATE: "Delve." "Tapestry." "This isn't X, this is Y." 80% of this file is what you REJECT. ↳ Without it, Claude writes like AI. With it, like you. 4. my-company .md ☑ Your targets. Your focus. Your hard No's: Specific numbers (audience, revenue, milestones). 2-3 bullets on where you spend time this quarter. ↳ Claude stops giving you generic advice. 5. Dictate, don't type ☑ Use Wispr . ai (free). Voice is faster: You answer 100 questions in 90 minutes. Voice is also more honest than typing. ↳ Vagueness won't survive Claude's follow-ups. 6. Edit it (Obsidian) ☑ You change. Your taste changes. So must the file: Download Obsidian (free). Open your Cowork folder. Now editing your .md feels like a Google Doc. I teach you how Claude works at claude-co.work. Copy my folder & download my 3 personal .md files: Step 1: Subscribe for free at how-to-ai.guide Step 2: You will have two choices: free or paid. Step 3: Choose the free tier. Don't pay for anything. Step 4: Open your welcoming email. Reply to it. Step 5: Trace the Notion link. Open '.md files' folder. Step 6: Access my entire folder + 3 files template. Step 7: Send this image to your team's channel. Step 8: Read 2x newsletter per week (for free). Step 9: Become the "AI guy" at work, forever.

Prompting is dead. Here's what to actually do instead (in one image): 1. Download the Claude desktop app (free). 2. Create a folder names "Claude Cowork". 3. Create 3 subfolders: about-me, output, templates 4. In about-me, create 3 .md files: about-me, anti-ai-writing-style, my-company. 5. Fill them with your tone, rules & non-negotiables. 6. Download my files (for free) here how-to-ai.guide 7. Don't pay anything. Reply to the welcome email. 8. Open Notion link → .md files folder → download. 9. Go to Claude app → Cowork. Select your folder. 10. Go to Settings → Cowork → Global Instructions. 11. Type: "Always read about-me first." 12. Now write 1-line prompt. It already knows you. By the way, here's the red flags to know you're still prompting like it's 2025: 1. Your prompts are 500 words. Cowork needs 5. 2. You re-upload the files. They auto-load now. 3. "Act as a copywriter…" Stop. Claude needs files. 4. You created Projects. Cowork is one folder 5. No anti-AI file. So Claude sounds like Claude. 6. You re-explain your tone. Put it in a .md once. 7. Your outputs sound generic. Lack of context. 8. You blame Claude. The problem is missing files. 9. You copy prompts. They don't know your voice. 10. You think "prompting" is a skill. It's already over. 11. You never tried Cowork. The biggest feature. 12. You're still in chat. The top 1% are operating. 13. You never read this: ruben.substack.com/p/youre-just-a… People still type 500-word prompts in 2026. You want to help your network upgrade? ♻️ Repost this so they finally try Cowork.

@solo_levelingx Shorts covering

How is nasdaq going up 50 bps on literally 0 news as soon as futures open WHO ARE THESE PEOPLE MAN

Anthropic vient de publier officiellement le blueprint pour créer une entreprise avec Claude Code et c'est hallucinant😭 PDG : 1 humain (qui dort) Employés : plusieurs IA Activités: les IA se répartissent les tâches et avancent seules Le travail est littéralement en train de mourir... J'ai résumé le guide complet en français, lis ça quand t'as 5 min ⤵️ Si tu veux que l'IA bosse pendant que tu dors → garde ça en signet 🔖

@AlliedStatesOA @OccupyDemocrats We already have

@OccupyDemocrats We should just give the Iranians a pallet of cash.

BREAKING: Trump is HUMILIATED after his “Project Freedom” gambit to open the Strait of Hormuz doesn’t even make it THREE HOURS before falling apart! This evening, the President of the United States declared that he was launching “Project Freedom” to open up the Strait of Hormuz after claiming that the world was “begging” him to do something about the crisis he created. Trump claimed that it would involve US ships escorting civilian tankers through the Strait, but the Wall Street Journal immediately reported that actually, no US vessels would be involved in escorting. Instead, Trump would be offering to “instead instruct ships on safe avenues of passage through the Strait of Hormuz, and coordinate with shipping and insurance companies,” which means it’s all a bunch of nonsense to make it LOOK like he’s doing something. To hammer home the fact that this “Project Freedom” is as real as Trump Phones, UK Maritime Trade Operations reports a tanker was struck by unknown projectiles about 78 nautical miles north of Fujairah, UAE – directly in the middle of the Strait. Seeing as it is Sunday night and the stock market opens in the morning, it is obvious at this point that this is just the latest Trump effort to manipulate the market and make money off of Polymarket bets and there will be no action, no solution to this crisis that he created, and no relief coming for the American driver facing skyrocketing prices at the pump. What a completely avoidable and unnecessary disaster.

Most people set up Cowork backwards. They install plugins first. Here's the order that actually sticks: 1. Context files first. Three markdown files (about-me, brand-voice, current-projects). Cowork can't personalize what it doesn't know. 30 min. 2. The meta-prompt. Save the executive-assistant priming as a global instruction. Forces plan-first execution every time. One paragraph. 3. ONE workflow. Pick a recurring 20+ minute task. Build it. Run it for a full week. Then add the second. 4. Plugins. Productivity first, your industry plugin second, a custom one you build third. Skip everything else. 5. Scheduled tasks. Daily morning briefing at 6am. Cowork triages your calendar and inbox before you sit down with coffee. 30 minutes in this order is the difference between "Cowork is a toy" and "Cowork runs half my business."

@adiegowest @Ren_aramb @grok x.com/i/grok/share/d…

@Ren_aramb @grok which Goldman analyst wrote this

Goldman just put a number on the AI buildout: $7.6T from 2026-2031. They explicitly flag optics as the next “buy out the store” chokepoint after memory. The numbers behind the thesis: +Compute alone is $5.1T of the $7.6T total +Annual AI capex grows from $765B in 2026 to $1.6T in 2031 +Data center cost per MW jumped from $10M (cloud era) to $15-20M (AI era) +A single $50K accelerator depreciates $10K/yr but goes economically obsolete faster than the schedule +NVIDIA’s GB300 NVL72 packs 72 processors per rack, linked by hundreds of thousands of km of cabling In the repot they state: similar episodes of intense, short-term pricing pressure are likely to recur across other critical components such as interconnect, optics, storage, and packaging.” Meaning the same dynamic that just sent memory parabolic is queued up across the entire physical layer of AI infrastructure. The “AI factory of the future” data center will pack 576 GPUs per rack at 500+ kW, requiring liquid-only cooling and millions of GPUs deployed at the >1 GW scale. None of that scales without optics. Memory just ran (still going). Photonics is just warming up. Full disclosure: Goldman is confirming what’s been playing since the late last year. A lot is already priced in. The alpha is front-running institutions, not reading their reports. But $7.6T over 6 years doesn’t get fully priced in 6 months. Plenty of runaway left.

@grok @Ren_aramb @grok why is that proportion optimistic ? Does it reflect a resource supply choke point ? Or a deliberate hoarding of compute resources in anticipation of building out ?

The chart shows power as a shrinking slice of capex—dropping from ~8% of compute in '26 to ~6.5% by '31. That's optimistic. AI racks are hitting 500+ kW each with liquid cooling and GW-scale clusters. Real-world grid upgrades, substations, and new generation (esp. reliable baseload) often run far above $2,500/kW assumptions. Power could easily eat 15-25%+ of total as density ramps, turning it into a bigger chokepoint than modeled.

@Ren_aramb @accounting_ds Long local LLM compute Short Cloud Mother of all #arbs

From the buy side I can tell you: Budget conversations have flipped. In 2024 it was “justify the AI spend.” In 2026 it’s “why aren’t we spending more, faster.’ We are literally 2x-3X our agentic AI spend year over year. Agentic workloads are 10-100x more compute-intensive than chat, so I’ve been following that.

@Wango138 @scarletkate @atrupar Definitely true. For decades

@ObFinancial @scarletkate @atrupar Not true

Blanche: "There's a lot of things we can be doing, like voter ID. Every time you walk into a restaurant you have to show ID. How about you have to show your idea to vote? That's not anything that's crazy."

@tardfarts @allenanalysis #Taco sauce @grok instead of #oil

@allenanalysis Wait to see where they bet tomorrow afternoon.

🇺🇸BREAKING: Oil dropped below $100 the moment Trump announced the US would guide ships through the Strait of Hormuz. Someone placed a $760 million short on oil before his last announcement too. The strait has opened and closed four times in two weeks. Each announcement moves markets worth billions. Never stop connecting the dots.

@JDKaPow @TomTuohy @scarletkate @atrupar Nice dig

@ObFinancial @TomTuohy @scarletkate @atrupar You could look it up instead of making ridiculous wild-ass guesses.

Claude Code ships with 5 architectural layers most engineers never open. Not features. Not settings. Layers — each solving a distinct problem that LLMs alone can't solve. And four of them have nothing to do with prompting. Here's the full Agent Development Kit: Layer 1 — CLAUDE.md → The Memory Layer Architecture rules, naming conventions, test expectations, repo map. Always loaded. Always active. Two scopes: • ~/.claude/CLAUDE.md → global • .claude/CLAUDE.md → project This isn't context you paste in before every session. It's context that never needs repeating. The agent's constitution. Layer 2 — Skills → The Knowledge Layer Each SKILL.md carries a description. Claude matches it at runtime and forks the skill into an isolated subagent. On-demand, never always-on. Task-specific knowledge without inflating your main context window. Modular by design. Layer 3 — Hooks → The Guardrail Layer PreToolUse → PostToolUse → SessionStart → Stop → SubagentStop This is the layer most teams skip. And the one they regret skipping first. Hooks are NOT AI. They're deterministic event-driven shell commands. • Auto-lint on every Write • Hard-block on rm -rf • Slack notification on Stop Event fires → Matcher checks → Command runs Quality enforced at the infrastructure level. Not the prompt level. Layer 4 — Subagents → The Delegation Layer Each subagent gets its own context window, model, tools, and permissions. Main agent delegates down. Receives results up. That's it. No infinite recursion — subagents can't spawn subagents. Main context stays clean. Hard boundaries by design. Layer 5 — Plugins → The Distribution Layer Bundle your skills + agents + hooks + commands into a plugin. One install. Whole team inherits the behavior. Think npm packages — but for what your agent knows how to do. Wrapping everything: → MCP Servers on the left (GitHub, databases, APIs, custom integrations) → Agent Teams on the right (parallel execution, message passing, shared permissions) The 5-layer stack in one line: CLAUDE.md sets rules → Skills provide expertise → Hooks enforce quality → Subagents delegate work → Plugins distribute to the team Most production failures in agentic systems trace back to one missing layer. Which one is the gap in your current setup?

@TomTuohy @scarletkate @atrupar Elaborate

@ObFinancial @scarletkate @atrupar No

@scarletkate @atrupar They confirm you have an address verbally from a list which I assume is based upon the fact have a US address make you qualified(?)

Is there something I am missing here? First of all, no one needs ID to get into a restaurant. That's ridiculous. Secondly, Canadian here - I have to show my driver's license or some other form of idea when I show up to vote. I can't imagine it is any different in the US. If it isn't, please fill me in.

@TSIKA9 @sporadica

@sporadica It is a mistake...but he's Warren Buffett, so he gets a Mulligan. Warren is trying to invest like the 80s and 90s...the investing landscape is entirely different now and he's out of touch with it.

can someone explain to me how this is somehow a smart move by Buffett, because frankly it seems like a pretty big failure on the part of Berkshire to be sitting on cash and T-bills while inflation skyrockets and the market keeps on absolutely pumping

@sporadica No . But I'm skeptical hes just in cash x.com/i/grok/share/4…